Solubility of Docosane in Heptane - Journal of Chemical

determined in the range of (277.25−317.65) K. The effect of ultrasonic waves in ... Citation data is made available by participants in Crossref'...
0 downloads 0 Views 51KB Size
1516

J. Chem. Eng. Data 2001, 46, 1516-1519

Solubility of Docosane in Heptane Karla C. M. Alves and Rodrigo Condotta* Department of Chemical Engineering, Federal University of Sao Carlos, P.O. Box 676, CEP 13565-905, Sao Carlos, SP, Brazil

Marco Giulietti† Division of Chemistry, Institute of Technology Research, P.O. Box 0141, CEP 01064-970, Sao Paulo, SP, Brazil

The solubility of docosane in heptane has been determined in the range of (277.25-317.65) K. The effect of ultrasonic waves in the solubility was studied. The results can be fitted with high accuracy to the relationship log10(x1) ) A + B/T + C log10(T). It was found out that the ultrasound did not affect the system solubility and the system behaved as an ideal solution.

Introduction Solubility data for a solute-solvent system are the starting point to determine or estimate other crystallization parameters, such as crystal nucleation and growth. Few data of solubility of long-chain paraffins in alkanes of low molecular weight are available in the literature.1-5 These data are relevant to the petroleum chemistry field, and they are also necessary to crystallization kinetics study. The solubility of a long-chain hydrocarbon in an organic solvent depends not only on the hydrocarbon chain length and the solvent molecular structure but also on the attractive forces between these molecules. Therefore, it is very difficult to predict if the system will behave as an ideal or nonideal mixture. The use of ultrasound technology in commercial applications has been increasing significantly in recent years. There is evidence that ultrasound can influence various chemical processes, including crystallization.1,2,6-8 Experimental Procedure Materials. The solvent was heptane purchased from Anidrol Prods. Quı´micos Ltd. with a stated boiling point of 371.45 K. The docosane (n-alkane C22), melting point 316.9 K, was purchased from Alfa Aesar (Johnson Matthey Co.). The purity of both substances was 99 mol %; therefore, they were used without further purification. Method of Measurement. Solutions of docosane in heptane of known concentration were prepared by mass; pure components were placed in an appropriate crystallizing apparatus (Figure 1), comprising an 80 cm3 glass reactor with an outer jacket, through which a mixture of water and ethylene glycol from a thermostat was circulated, a glass four-blade stirrer (500 rpm), and two ultrasonic cleaning baths filled with water. One of the baths operated with a frequency of 25 kHz and a 50 W power supply, and the other operated with a frequency of 40 kHz and a 30 W power supply. The glass reactor was closed to * Author to whom correspondence should be addressed (e-mail [email protected]; fax +55-11-3767-4052). † E-mail: [email protected].

Figure 1. Experimental apparatus.

prevent the solvent evaporation, and it was maintained submerged in the ultrasonic bath. The temperature of the solution was continuously recorded using a mercury thermometer with an accuracy of (0.05 K. The solubility was determined according to Ny´vlt’s methodology,9 by which the solution was preheated until all crystals were dissolved, then the solution was cooled slowly at three different constant rates of (6.0, 18.0, and 30.0) K/h to a temperature at which a sufficient number of tiny crystals were formed, and, finally, the temperature was slowly raised at the rate of 0.6 K/h until the last crystal was dissolved. This temperature was considered to be the saturation temperature of solution, Teq. The temperature could be determined with considerable precision by careful observation through magnifying glasses when the solution was illuminated with a powerful light at right angles to the observation axis. The procedure was repeated three times for each cooling rate. The deviation during reproducibility of experiments was just about 0.1 K.1,2,9-11 The effect of ultrasonic waves on the solubility of docosane in heptane was also determined. The experiments were carried out in the same form. The ultrasonic generator was allowed to operate continuously during the experiments.

10.1021/je010032h CCC: $20.00 © 2001 American Chemical Society Published on Web 10/09/2001

Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001 1517 Table 1. UNIFAC Parameters

Table 2. Solubility of Docosane (1) in Heptane (2)

group

subgroup

R

Q

T/K

x1

T/K

x1

T/K

x1

(1) CH2 (1) CH2

(1) CH3(2) -CH2-

0.9011 0.6744

0.848 0.540

277.25 283.15 286.25 289.30 290.95

0.0225 0.0440 0.0647 0.0844 0.1033

295.35 297.35 298.75 301.45 302.35

0.1556 0.1873 0.2166 0.2694 0.2932

304.75 305.55 307.65

0.3561 0.3922 0.4668

Prediction The solubility of any solid substance in an ideal system at constant pressure follows the relationship4,5,12

(

T/K

)

∆Hf1 1 1 ln x1 ) R Tm T

(1)

where ∆Hf1 is the enthalpy of fusion of the solute at its melting point (Tm) and x1 is the solute mole fraction. For the substances with a first-order transition point we have, for T < Ttr5,6,12

(

)

(

)

∆Hf1 1 ∆Hf1 + ∆Htr1 1 1 1 + (2) ln x1 ) R Tm Ttr R Ttr T where ∆Htr1 is the enthalpy on transition of the solid and Ttr is the pure solid transition temperature. For a nonideal system, eq 3 provides a reasonable method for predicting the saturation mole fraction of a solid in a liquid:3,4,12

(

[( )

)

∆Hf1 Tm ∆Cp1 Tm T ln γ1x1 ) -1 + -1 ln + RTm T R Tm T

]

)

Method.4,13,14

]

(

expt 1

expt 2

expt 3

0.0225 0.1873 0.4668

277.25 297.40 307.60

277.30 297.30 307.70

277.20 297.35 307.65

activity coefficient can be predict by the ScatchardHildebrand relation

RT ln γ1 ) υ1(δ1 - δ2)2φ22

(5)

where υ1 is the molar volume of pure solute, δ1 and δ2 are the solubility parameters of the solvent and solute, respectively, and φ2 is the volume fraction of the solvent, calculated by

φ2 ) x2υ2/(x1υ1 + x2υ2)

(6)

(3)

∆Hf1 Tm ln γ1x1 ) -1 RTm T Tm ∆Cp1 ∆Cp1 Tm ∆Htr1 Ttr -1 + - 1 (4) ln + R T R T RTtr T

[

x1

where xi denotes the mole fraction. The solubility parameter δi defined by

In eq 3 γ1 is the liquid-phase activity coefficient of the solute and ∆Cp1 is the change of the heat capacity of the pure solute on fusion (∆Cp1 ) ∆Cp1(liquid) - ∆Cp1(solid)). For an ideal system, eq 3 could be used with the parameter γ1 equal to 1. When a phase transition takes place between T and Tm, eq 3 should be modified to include the effect of the transition point:3-5

( [ ( )]

Table 3. Solubility of Docosane (1) in Heptane (2) from Ultrasound Measurements

)

UNIFAC This is based on the assumption that a physical property of a compound is the sum of contributions made by all of the functional groups in the molecule. The molecular activity coefficient is separated into two parts, combinatorial and residual. The former is due to differences in the shape and size of the molecules in the mixture, whereas the latter arises from interactions between structural groups. Both sizes of the functional groups and the interaction areas are taken from pure component molecular structure data. The program used in this paper was UNIFAC Activity Coefficient Calculator version 3.013 with procedure 6A of 1986, with which the docosane was composed by 2 molecules of subgroup CH3 and 20 molecules of CH2 and heptane was composed by 2 and 5 molecules of subgroups CH3 and CH2, respectively. The values of parameters R and Q were extracted from the program and are shown in Table 1. More details are given by Fredenslund.14 Regular Solution Theory.4,5,12 For nonpolar solvent and solute molecules using only pure component data the

1/2 δi ) (∆Uvap i /υi)

(7)

where ∆Uvap is the molar energy of vaporization for the i is the molar enthalpy of pure component i and ∆Hvap i vaporization of pure component i at temperature T. The could be calculated from the molar envalue of ∆Uvap i thalpy of vaporization or by the difference between the molar enthalpies of sublimation and fusion of the component i, both at the same arbitrary temperature T.

∆Uvap ) ∆Hvap - RT ) ∆Hsub - ∆Hfus i i i i - RT

(8)

Those two methods were used to estimate the activity coefficient in eq 4 for a nonideal system. Results The solubility of docosane in heptane was measured in the temperature interval (277.25-307.65) K. The results are given in Table 2, where T represents the saturation temperature of the solution in which the docosane has a mole fraction x1. It was found that the ultrasound did not influence docosane solubility in heptane solutions. The values for the system solubility under ultrasound conditions are shown in Table 3. The measured solubility in the region of temperature and concentration studied is well represented by the equation proposed by Ny´vlt9

log10 x1 ) A +

B + C log10 T T

(9)

with A ) 19.0175, B ) -3980.1740, and C ) -2.5733. This equation was used to estimate the amount of solids produced in the batch experiments because direct measurement was not practical. This equation can also be used

1518

Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001

Figure 2. Solubility of docosane (1) in heptane (2).

Figure 3. Evaluated solubility of docosane (1) in heptane (2).

Table 4. Some Activity Coefficients for the Docosane (1)-Heptane (2) System γ1 calculated by T/K

UNIFAC method

regular solution theory

290.95 301.45 305.55

0.635 0.802 0.875

1.095 1.041 1.021

to evaluate the initial concentration x1 for batch experiments starting from a solution saturated at a temperature Teq and cooled to a temperature Tf.1,2 Figure 2 shows the experimental data and the curve fitted by the model of eq 9. The mean relative difference observed between the experimental data and the model was 0.52%. Discussion The experimental values of the system solubility were compared with two evaluations of solubility data by eq 2 and 4 for ideal solution and by eq 4 using the UNIFAC Method and Regular Solution Theory to estimate the activity coefficient. In both case, the thermodynamic and physical property data for docosane (1) and heptane (2) have been taken from the NIST Standard Reference Data Program,15 and they were near those from Knlaz’s work3 (MW1 ) 310.60 g‚mol-1; F1 ) 0.7644 g/cm-3; MW2 ) 100.20 g‚mol-1; F2 ) 0.684 g‚cm-3; ∆Hf1 ) 47801 J‚mol-1; Tm ) 316.8 K; ∆Htr1 ) 28605 J‚mol-1; Ttr ) 315.5; ∆Hsub ) 1 vap -1; Tvap ) 391 K; ∆H ) 31770 J‚mol 172600 J‚mol-1; Tsub 1 2 2 ) 371.6 K). The activity coefficients calculated by UNIFAC were smaller than 1 (ideality), but the ones calculated by Regular Solution Theory were approximately equal to unity. These values are shown in Table 4. The evaluated solubility of the system by these methods is shown in Figures 3 and 4. The group contribution methods are very approximate, which explains the low values for the activity coefficient calculated by the UNIFAC Method. They are based on the assumption that a functional group within a molecule acts independently on other groups and that its contribution to a physical property of any molecule is the same. The method gives better results with increasing difference between the main groups.4 In this system, the two components (heptane and docosane) belongs to the same group (group 1). The mean relative difference between the experimental data and the model given by eq 2 (considering only the

Figure 4. Evaluated solubility of the system in logarithm scale.

phase transition) was 2.79% and 1.86% between experimental data and eq 4 (considering the phase transition and the heat capacity on fusion) for an ideal solution (γ ) 1). When the system was studied as a nonideal solution (assuming γ * 1), deviations between the experimental and calculated data were greater. For the Regular Solution Theory the results deviation was 2.69% and 3.80% for UNIFAC method. Therefore, the system can be interpreted as an ideal solution. Literature Cited (1) Alves, K. C. M. Intensification of Paraffin Crystallization in Organic Solvents with Ultrasound. M.Sc. Thesis, Federal University of Sao Carlos (UFSCar), Sao Carlos, Brazil, 1999. (2) Martinez, K. C. L.; Condotta, R.; Giulietti, M. Intensification of Paraffin Crystallization with Ultrasound. Presented at the 14th International Symposium on Ind. Crystallization, Cambridge, U.K., 1999. (3) Kniaz, K. Solubility of Normal-Docosane in Normal-Hexane and Cyclohexane. J. Chem. Eng. Data 1991, 36, 471-472. (4) Brecevic, L.; Garside, J. Solubilities of Tetracosane in Hydrocarbon Solvents. J. Chem. Eng. Data 1993, 38, 598-601. (5) Madsen, H. E. L.; Boistelle, R. Solubility of Long-Chain NormalParaffins in Pentane and Heptane. J. Chem. Soc., Faraday Trans. 1 1976, 72, 1078-1081. (6) Kruus, P., et al. Effects of ultrasound on the cementation of Cobalt Zinc. Ultrasonics 1991, 29, 370-375. (7) Suslick, K. S., et al. On the Origin of Sonoluminescence and Sonochemistry. Ultrasonics 1990, 28, 280-290. (8) Ny´vlt, J.; Za´cek, S. Effect of Ultrasonics on Agglomeration. Cryst. Res. Technol. 1995, 30, 1055-1063.

Journal of Chemical and Engineering Data, Vol. 46, No. 6, 2001 1519 (9) Ny´vlt, J.; So¨nhel, O.; Matuchova´, M.; Broul, M. The Kinetics of Industrial Crystallization; Academia: Prague, Czech Republic, 1985. (10) Ny´vlt, J. Solid-Liquid Phase Equilibria; Elsevier: Amsterdam, The Netherlands, 1977. (11) Mullin, J. W. Crystallization, 3rd ed.; Butterworth-Heinemann: London, U.K., 1993. (12) Prausnitz, J. M. Molecular Thermodynamics of Fluid-Phase Equilibria, 2nd ed.; Prentice-Hall: Englewood Cliffs, NJ, 1985. (13) http://www.chem.eng.usyd.edu.au/pgrad/bruce/unifacal/ calc.htm.

(14) Fredenslund, A.; Gmehling, J.; Rasmussen, P. Vapor-Liquid Equilibria using UNIFAC; Elsevier: Amesterdam, The Netherlands, 1977. (15) http://webbook.nist.gov. Received for review January 29, 2001. Accepted August 2, 2001. We are grateful to FAPESP (Fundation to Support Research in Sao Paulo State) for financing this project.

JE010032H